& cplSiteName &

Dr. Lawrence Roberts

Light Reading
Interview
Light Reading
1/25/2001
50%
50%

Light Reading: What else can you tell us about your product?

Larry Roberts: Reliability is another thing we are working on. Service providers need equipment with carrier-grade reliability. Equipment that can stand up to upgrades -- so that they can perform software upgrades and fix failed boards and offer uninterrupted service for five or ten years, or until they see a return on their investment.

And right now they don’t have that. The feedback that we’ve gotten from all the service providers is that products should have hitless software upgrades and hitless cutovers for BGP [border gateway protocol] when a failure occurs. And it’s the hardest thing to do – especially for BGP. None of today’s router vendors have succeeded in doing it.

What does that mean? It means that in any network today carriers have to have complete duplication of the [routing] hardware. They have two have two duplicate pieces of equipment sitting next to each other [for redundancy]. They’re both live; they’re both working, but at half their capacity.

And that’s not the only problem. The lines are working below capacity because of QOS inefficiencies. If you look at networks today, some of them are running at 15 percent to 30 percent utilization.

And they don’t have any way to manage load balancing on the network, so they have to do that manually. They’re having to scale [the number of engineers they hire] linearly with the traffic. This is a major cost for them.

So to sum up, the first network cost is to light the fiber. Then the cost of the switching equipment. And then the operational costs -- of the people that run it. And right now these guys are sitting there with data accounting for 80 percent of their traffic, but only 20 percent of revenue. And data will be 90 percent of their traffic next year, so they better make IP profitable within a year or they are going to go under.

Light Reading: And Caspian will help them do that?

Roberts: Yes. Only a slight change in the cost of the switch can bring them to profitability. And then the IP SLAs [service level agreements] can make them very profitable.

 
(14)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
Page 1 / 2   >   >>
mahadeva
50%
50%
mahadeva,
User Rank: Light Beer
12/4/2012 | 8:59:05 PM
re: Dr. Lawrence Roberts
Is that a misprint??? Who in the industry is working on petabit switches? I thought there were only a few companies working in the 1 to 10 terabit range.
pablo
50%
50%
pablo,
User Rank: Light Beer
12/4/2012 | 8:59:04 PM
re: Dr. Lawrence Roberts
" .. I thought there were only a few companies working in the 1 to 10 terabit range .."

Which, of course, is 0.01 petabits... :-) That is yesterday's news, zettabits is where it's at... :-)
bryan_gregory
50%
50%
bryan_gregory,
User Rank: Light Beer
12/4/2012 | 8:59:02 PM
re: Dr. Lawrence Roberts
Mahadeva,

Check out Hyperchip's web site for PetaBit Switch information. I guess Terabit's will be old-hat in a couple of years :-)

WWW.HYPERCHIP.COM

Regards,
Bryan-
gea
50%
50%
gea,
User Rank: Light Beer
12/4/2012 | 8:58:56 PM
re: Dr. Lawrence Roberts
I'd place any amount of money I know exactly what Caspian is doing, and remember you heard it from "gea" first. Note that Daniel Blumenthal from UCSB has taken a leave to join Caspian.

What Caspian is going to try to do with optics is bascially the same thing that Ipsilon did with an ATM switch (converting a switch into a router by identifying switchable flows). With the Caspian switch, any time a flow is big enough to merit siwtching in the optical domain (ie, MEMs), then the electronic matrix will tag the flow and dump it down into the optical matrix.

In that sense, it is easy to scale to petabits, because once you have a wavelength flow, you've got 10Gb/s. Enough wavelength flows, and you're switching Petabits.

chechaco
50%
50%
chechaco,
User Rank: Light Beer
12/4/2012 | 8:58:53 PM
re: Dr. Lawrence Roberts
Do you remember what happened to Ipsilon? Dynamic identification of flows does not scale well. That's why people use CR-LDP/RSVP-TE to set up connections in IP environment.
redface
50%
50%
redface,
User Rank: Light Beer
12/4/2012 | 8:58:40 PM
re: Dr. Lawrence Roberts
Hi Gea, Daniel Blumenthal joined Calient, not Caspian.

What you say about Calient is interesting, please feed us more. Are they doing any dynamic wavelength conversion?
netskeptic
50%
50%
netskeptic,
User Rank: Light Beer
12/4/2012 | 8:58:39 PM
re: Dr. Lawrence Roberts
"Do you remember what happened to Ipsilon? Dynamic identification of flows does not scale well. That's why people use CR-LDP/RSVP-TE to set up connections in IP environment."

I would rather say that people have a hope that MPLS and RSVP would work one day. In many respects it vividly reminds me the hopes pinned on ATM in its early days.
bobilll
50%
50%
bobilll,
User Rank: Light Beer
12/4/2012 | 8:58:22 PM
re: Dr. Lawrence Roberts
What do people think of this statement?

"Roberts: I have a similar problem with it. And I have another problem. Once you successfully do photonic switching, itGÇÖs so cheap, where will [photonic switch manufacturers] get the revenue? LetGÇÖs just count the ports and how many there are going to be and what the market size is going to be. ItGÇÖs actually in the hundreds of millions at the best. "

The photonic switching companies carry hefty valuations based in large measure on perceived potential for exceptional revenue growth. Is the perception incorrect?-- will the revenue dry up because the market is smaller than thought?

Big A
50%
50%
Big A,
User Rank: Light Beer
12/4/2012 | 8:57:42 PM
re: Dr. Lawrence Roberts
I agree with Roberts: There is little value add with core transport. There will be little intelligence in these boxes. Making the optical device to switch might be better, but actually moving the photons won't be a huge market.
Genuine
50%
50%
Genuine,
User Rank: Light Beer
12/4/2012 | 8:56:50 PM
re: Dr. Lawrence Roberts
When I set out to read this interview, like other articles I am spending time with, what I was hoping for is what will make IP profitable? What keeps business owners up at night? The Internet? Does business find themselves lacking in bandwidth to accomplish their revenue goals with the internet? How much business is coming from the Internet anyway? What are the goals beside moving data and e-mail? Do they think they will be left out of the "New Economy" if they do not get a three demensional/fully interactive Web site? Isn't that what sales people are? What is the "New Economy"-Cisco thinks it is voice over IP. But then again, isn't that taking way the only considerable profitablility (economy) of a telecommunications/ISP? I guess the question boils down to, What markets are you creating with your products? Please do not tell me the "New Economy". Tell me your vision.
Page 1 / 2   >   >>
Featured Video
Flash Poll
Upcoming Live Events
December 4-6, 2018, Lisbon, Portugal
March 12-14, 2019, Denver, Colorado
April 2, 2019, New York, New York
April 8, 2019, Las Vegas, Nevada
May 6-8, 2019, Denver, Colorado
All Upcoming Live Events